The Rocknest sand shadow analyzed by the Curiosity rover on Mars was similar to coarse-grained ripples analyzed by previous rovers. It consisted of an upper layer of very coarse sand grains armoring the surface, underlain by finer grains. Analysis found the sand was around 55% crystalline material of basaltic composition and 45% amorphous iron-rich glass. This amorphous component contained the volatiles detected and was similar to soils analyzed at other Mars sites, implying the materials were locally derived from similar basaltic sources globally on Mars.

1. DOI: 10.1126/science.1239505
, (2013);341Science
et al.D. F. Blake
Rocknest Sand Shadow
Curiosity at Gale Crater, Mars: Characterization and Analysis of the
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by
, you can order high-quality copies for yourIf you wish to distribute this article to others
here.following the guidelines
can be obtained byPermission to republish or repurpose articles or portions of articles
):September 29, 2013www.sciencemag.org (this information is current as of
The following resources related to this article are available online at
http://www.sciencemag.org/content/341/6153/1239505.full.html
version of this article at:
including high-resolution figures, can be found in the onlineUpdated information and services,
http://www.sciencemag.org/content/suppl/2013/09/26/341.6153.1239505.DC1.html
can be found at:Supporting Online Material
http://www.sciencemag.org/content/341/6153/1239505.full.html#related
found at:
can berelated to this articleA list of selected additional articles on the Science Web sites
http://www.sciencemag.org/content/341/6153/1239505.full.html#ref-list-1
, 11 of which can be accessed free:cites 34 articlesThis article
http://www.sciencemag.org/content/341/6153/1239505.full.html#related-urls
3 articles hosted by HighWire Press; see:cited byThis article has been
registered trademark of AAAS.
is aScience2013 by the American Association for the Advancement of Science; all rights reserved. The title
CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
onSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfromonSeptember29,2013www.sciencemag.orgDownloadedfrom

2. Curiosity at Gale Crater, Mars:
Characterization and Analysis
of the Rocknest Sand Shadow
D. F. Blake,1
* R. V. Morris,2
G. Kocurek,3
S. M. Morrison,4
R. T. Downs,4
D. Bish,5
D. W. Ming,2
K. S. Edgett,6
D. Rubin,7
† W. Goetz,8
M. B. Madsen,9
R. Sullivan,10
R. Gellert,11
I. Campbell,11
A. H. Treiman,12
S. M. McLennan,13
A. S. Yen,14
J. Grotzinger,15
D. T. Vaniman,16
S. J. Chipera,17
C. N. Achilles,2
E. B. Rampe,2
D. Sumner,18
P.-Y. Meslin,19
S. Maurice,19
O. Forni,19
O. Gasnault,19
M. Fisk,20
M. Schmidt,21
P. Mahaffy,22
L. A. Leshin,23
D. Glavin,22
A. Steele,24
C. Freissinet,22
R. Navarro-González,25
R. A. Yingst,16
L. C. Kah,26
N. Bridges,27
K. W. Lewis,28
T. F. Bristow,1
J. D. Farmer,29
J. A. Crisp,14
E. M. Stolper,15
D. J. Des Marais,1
P. Sarrazin,30
MSL Science Team‡
The Rocknest aeolian deposit is similar to aeolian features analyzed by the Mars Exploration
Rovers (MERs) Spirit and Opportunity. The fraction of sand <150 micrometers in size contains
~55% crystalline material consistent with a basaltic heritage and ~45% x-ray amorphous material.
The amorphous component of Rocknest is iron-rich and silicon-poor and is the host of the volatiles
(water, oxygen, sulfur dioxide, carbon dioxide, and chlorine) detected by the Sample Analysis at
Mars instrument and of the fine-grained nanophase oxide component first described from
basaltic soils analyzed by MERs. The similarity between soils and aeolian materials analyzed at
Gusev Crater, Meridiani Planum, and Gale Crater implies locally sourced, globally similar
basaltic materials or globally and regionally sourced basaltic components deposited locally at
all three locations.
T
he Mars Science Laboratory (MSL) rover
Curiosity began exploring the surface of
Mars on 6 August 2012 (universal time co-
ordinated); until 13 September 2012, it conducted
an initial engineering checkout of its mobility sys-
tem, arm, and science instruments. Curiosity spent
sols 57 to 100 (1) at a location named Rocknest,
collecting and processing five scoops of loose, un-
consolidated materials extracted from an aeolian
sand shadow (2).
Five scoops of material from the Rocknest
sand shadow were individually collected and
sieved (<150 mm) by the Sample Acquisition,
Sample Processing and Handling–Collection
and Handling for In situ Martian Rock Analysis
(SA/SPaH-CHIMRA) instrument (3). Scoops 1 and
2 were processed by CHIMRA and discarded
to reduce (by entrainment and dilution) any ter-
restrial organic contamination that may have
remained after a thorough cleaning on Earth (4)
and to coat and passivate the interior surfaces of
the collection device with Mars dust. Portions
(40 to 50 mg) of scoops 3 and 4 were delivered
to the Chemistry and Mineralogy (CheMin) in-
strument (5) and the “observation tray,” a 7.5-cm-
diameter flat Ti-metal surface used for imaging
and analyzing scooped and sieved material with
Curiosity’s arm and mast instruments. Portions of
scoop 5 were delivered to both CheMin and the
Sample Analysis at Mars (SAM) quadrupole mass
spectrometer/gas chromatograph/tunable laser
spectrometer suite of instruments (6).
We describe the physical sedimentology of
Rocknest and suggest possible sources for the
material making up the sand shadow. We use
Alpha-Particle X-ray Spectrometer (APXS) and
CheMin data to determine the amounts and chem-
istry of the crystalline and amorphous components
of the sand shadow and compare these results with
global soil measurements from the Mars Explora-
tion Rovers (MERs) and to basaltic martian mete-
orites analyzed on Earth.
Results
Description and Interpretation of the
Rocknest Sand Shadow
The Rocknest sand shadow (7) is an accumula-
tion of wind-blown sediment deposited in the
lower-velocity lee of an obstacle in the path of
the wind. The orientation of the sand shadow in-
dicates that the constructive winds were from the
north. The surface is composed of dust-coated,
predominantly rounded, very coarse (1- to 2-mm)
sand grains (Fig. 1A). Trenches created during
the scooping show that these larger grains form
an armored surface ~2 to 3 mm in thickness (Fig.
1B). Beneath the armored surface, the bedform
interior consists of finer-grained material whose
size distribution extends through the resolution
limit of Mars Hand Lens Imager (MAHLI) im-
ages (~30 mm per pixel under the conditions of the
observation) (8). Because of CHIMRA’s 150-mm
sieve, the larger grains that armor the surface
could not be analyzed by CheMin.
Coarse sand grains that fell from the crust
into the scoop-troughs lost their dust coating
and show diversity in color, luster, and shape.
Among the grains are gray and red lithic frag-
ments, clear/translucent crystal fragments, and
spheroids with glassy luster (Fig. 1C). Some grains
showed bright glints in the martian sunlight,
suggesting specular reflections from mineral crys-
tal faces or cleavage surfaces [similar features
were observed by the optical microscope on board
the Mars Phoenix Lander (9)]. MAHLI images
of a sieved portion of material deposited on the
observation tray (3) showed a variety of particle
types from clear to colored to dark, angular to
spherical, and dull to glassy-lustered (Fig. 1D).
During the scooping process, fragments of the
armored surface were cohesive to the extent that
“rafts” of surface crust were laterally compressed
and displaced forward, and fragments of the crust
fell into the scoop hole as cohesive units (Fig. 1B).
The surface crust was also fractured and broken
into rafts during scuffing by rover wheels (a pro-
cess by which an excavation is made into the sub-
surface of unconsolidated regolith by rotating a
single rover wheel). Material beneath the crust
also had some cohesion, as shown by the over-
steep walls of the scoop scars (much greater than
the angle of repose and vertical in some cases).
The sand shadow has a discernible internal
structure. On the headwall and flanks of each
scoop trench, a lighter-tone layer is apparent
~1 cm beneath and parallel to the dune surface
(Fig. 1B). The origin of the layering is not un-
derstood, and three hypotheses are viable. First,
RESEARCH ARTICLE
1
National Aeronautics and Space Administration (NASA) Ames
Research Center, Moffett Field, CA 94035, USA. 2
NASA Johnson
Space Center, Houston, TX 77058, USA. 3
Department of Geolog-
ical Sciences, University of Texas, Austin, TX 78712, USA. 4
Depart-
ment of Geology, University of Arizona, Tucson, AZ 85721,
USA. 5
Department of Geological Sciences, Indiana University,
Bloomington, IN 47405, USA. 6
Malin Space Science Systems,
San Diego, CA 92191, USA. 7
U.S. Geological Survey, Santa Cruz,
CA 95060, USA. 8
Max-Planck-Institut für Sonnensystemforschung,
37191 Katlenburg-Lindau, Germany. 9
Niels Bohr Institute,
University of Copenhagen, 2100 Copenhagen, Denmark. 10
Center
forRadiophysicsandSpaceResearch,CornellUniversity,Ithaca,NY
14850, USA. 11
University of Guelf, Guelph, Ontario, N1G2W1,
Canada.12
LunarandPlanetaryInstitute,Houston,TX77058,USA.
13
State University of New York–Stony Brook, Stony Brook, NY
11790, USA. 14
Jet Propulsion Laboratory/California Institute of
Technology, Pasadena, CA 91109, USA. 15
California Institute of
Technology, Pasadena, CA 91125, USA. 16
Planetary Science
Institute,Tucson,AZ85719,USA.17
ChesapeakeEnergy,Oklahoma
City, OK 73102, USA. 18
University of California, Davis, CA 95616,
USA. 19
Institut de Recherche en Astrophysique et Planétologie
(IRAP), UPS-OMP-CNRS, 31028 Toulouse, France. 20
Oregon State
University, Corvallis, OR 97331, USA. 21
Finnish Meteorological
Institute, Fl-00101 Helsinki, Finland. 22
NASA Goddard Space
Flight Center, Greenbelt, MD 20771, USA. 23
Rensselaer Poly-
technic Institute, Troy, NY 12180, USA. 24
Geophysical Laboratory,
Carnegie Institution of Washington, Washington, DC 20015, USA.
25
University Nacional Autonóma de México, Ciudad Universitaria,
04510 México D.F. 04510, Mexico. 26
Department of Earth and
Planetary Sciences, University of Tennessee, Knoxville, TN 37996,
USA. 27
The Johns Hopkins University Applied Physics Labora-
tory, Laurel, MD 20723, USA. 28
Princeton University, Princeton,
NJ 08544, USA. 29
Arizona State University, Phoenix, AZ 85004,
USA. 30
SETI Institute, Mountain View, CA 94043, USA.
*Corresponding author. E-mail: david.blake@nasa.gov
†Present address: Department of Earth and Planetary Sciences,
University of California, Santa Cruz, CA 95064, USA.
‡MSL Science Team authors and affiliations are listed in the
supplementary materials.
www.sciencemag.org SCIENCE VOL 341 27 SEPTEMBER 2013 1239505-1

3. the layering may represent changes in bulk com-
position or grain size that occurred during dep-
osition. Second, the layering may be the result
of changes in oxidation state or other chemical
properties that occurred after deposition, in which
case the conformable nature of the banding and
the surface of the sand shadow reflect depth-
dependent postdepositional chemical processes.
Finally, the layering may represent zones richer
or poorer in light-toned dust, reflecting times of
lesser or greater sand accumulation relative to
the air-fall dust.
The aeolian bedform at Rocknest is quite sim-
ilar to coarse-grained ripples encountered at Gusev
by the MER Spirit (10, 11) and at Meridiani
Planum by the MER Opportunity (12, 13) in that
a coarse-grained, indurated, dust-coated surface
overlies an interior of markedly finer sediment.
Coarse-grained ripples on Earth typically consist
of a surface veneer of coarse grains and a finer-
grained interior (7, 14), and the martian bed-
forms have been considered analogous features
(13, 15). The spatial grain-size sorting within
coarse-grained ripples is thought to arise because
of the short grain excursion length of the coarse
grains traveling in creep and the much longer ex-
cursion length of finer saltating grains (16). With
ripple migration,coarse grains are recycled through
the bedform and become concentrated on the
ripple surface, where impacts from saltating grains
tend to buoy the grains upward.
Although the dynamics of sand shadows dif-
fer from those of coarse-grained ripples, and sand
shadows on Earth do not characteristically show
a coarse-grained surface, similar dynamics may
arise owing to the mix-load transport of grains in
creep and saltation. Alternate interpretations are
also possible. First, the coarse-grained surface
could represent a lag formed as winds deflated
finer grains. However, the paucity of coarse grains
within the interior indicates that an unreasonable
amount of deflation would have had to occur to
produce the veneer. Second, the coarse-grained
veneer could represent the terminal growth phase
of the bedform. Because the size of a sand shad-
ow is fixed by the upwind obstacle size (17),
once the terminal size is approached, the lower
wind speeds that characterize the wake and allow
for deposition of finer sediment are replaced by
wind speeds that approach the unmodified (pri-
mary) winds. At this point, there would be se-
lective deposition of coarse grains traveling in
creep, whereas finer saltating grains would by-
pass the bedform. Third, the sand shadow could
have formed largely by the more readily trans-
ported fine saltation load, but as the area became
depleted in finer grains, more of the residuum of
Fig. 1. The Rocknest sand shadow, where Cu-
riosity spent sols 57 to 100 conducting engi-
neering tests and science observations of the
material. (A) Mosaic of 55 MAHLI images show-
ing Curiosity parked on the east side of the Rocknest
sand shadow during the sampling campaign on sol
84. The location of each of the five scoops is indi-
cated. The inset is a portion of Mars Reconnaissance
Orbiter High Resolution Imaging Science Experiment
image ESP_028678_1755 showing the Rocknest
sand shadow as seen from about 282 km above
the ground. (B) MAHLI image of third scoop trench,
showing the dust-coated, indurated, armoring layer
of coarse and very coarse sand and underlying darker
finer sediment. (C) MAHLI image of Rocknest sand
shadow surface disrupted by the rover’s front left
wheel on sol 57. The larger grains came from the
armoring layer of coarse sand on the sand shadow
surface. (D) MAHLI image of a <150-mm sieved por-
tion from the third scoop; grains similar to those
delivered to the CheMin and SAM instruments, de-
livered to Curiosity’s Ti observation tray.
27 SEPTEMBER 2013 VOL 341 SCIENCE www.sciencemag.org1239505-2
Curiosity at Gale Crater

4. coarser grains would be incorporated into trans-
port, with the coarse-grained surface arising through
subsequent deflation.
None of these interpretations explains the gen-
eral absence of observed coarse grains in the in-
terior; the contrast in grain size between the surface
and the interior is more marked in the Rocknest
sand shadow and in some of the coarse-grained
ripples observed by MERs than in many Earth
examples. This may reflect the greater impact en-
ergy of saltating grains on Mars compared with
Earth and their ability to transport dispropor-
tionally larger grains in creep (18). Regarding the
apparent absence of interior coarse grains, the small
scooped areas may not be representative of the en-
tire bedform, and interior horizons of coarse grains
could easily have been bypassed. In addition, as seen
with coarse-grained ripples on Earth, the amount of
coarse sediment occurring in the interior varies
and decreases with the supply of coarse grains.
Regardless of the origin of the coarse-grained
surface, this armored surface would stabilize
the bedform during all but the strongest wind
events. In turn, the armored surface would allow
time for surface induration to develop, further sta-
bilizing the sand shadow. The similarity of the
armoring and induration of the sand shadow at
Rocknest to coarse-grained ripples encountered
by Spirit and Opportunity suggests that the pro-
cesses of grain transport and stabilization are
similar across equatorial Mars and that Mars’
winds (in recent eras) rarely were strong enough
to transport sand grains of 1- to 3-mm diam-
eter. To move the grains at the current atmo-
spheric pressure of 0.02 kg/m3
, the wind velocities
would need to be ~36 m/s (80 mph) and ~52 m/s
(116 mph), with and without saltation, respec-
tively. Under conditions of high obliquity, dur-
ing which time the atmospheric pressure could
increase to 0.04 kg/m3
, these values would de-
crease to 26 m/s (58 mph) and ~37 m/s (83 mph),
respectively (see Materials and Methods). The
potential antiquity of the Rocknest sand shadow
is highlighted by comparing it with granule ripples
on Meridiani Planum, where cratering postdates a
field of pristine granule ripples and the crater count
suggests an age of 50,000 to 200,000 years (19).
Mineralogy of the Rocknest Sand Shadow
Analysis and interpretation of the mineralogy of
the Rocknest sand shadow is given in Bish et al.
(20). Rocknest consists of both crystalline and
x-ray amorphous components. The crystalline
component is basaltic, composed of plagioclase
feldspar, forsteritic olivine, and the pyroxenes
augite and pigeonite (20). All of the minor phases
are consistent with a basaltic heritage, with the
exception of anhydrite and hematite. By constrain-
ing the compositions of the individual crystalline
phases on the basis of their measured unit-cell
parameters, the chemical compositions of the
minerals of Rocknest were determined (21, 22).
The crystalline component of Rocknest is
chemically and mineralogically similar to that
inferred for martian basalts across the planet
and many of the basalts found in martian me-
teorites (Table 1) and, apart from somewhat
lower Fe and K, broadly similar to estimates of
the average martian crust (23). These basalts all
contain (or have chemical compositions consist-
ent with) the minerals olivine, augite, pigeonite,
and plagioclase feldspar. The mineral propor-
tions of the crystalline component of Rocknest
are virtually identical to those calculated for the
unaltered Adirondack class basalts from Gusev
Crater (CIPW normative mineralogy from their
APXS analyses) (Table 1) (24, 25). Chemically,
the mafic minerals of the Rocknest sediment (oli-
vine, augite, and pigeonite) are all consistent with
high-temperature chemical equilibria among Ca,
Fe, and Mg at 1050 T 75°C (Fig. 2). This con-
sistency with chemical equilibria suggests, but
does not prove, that these minerals and the plagio-
clase feldspar all derived from a common basaltic
source rock, which was broken down into indi-
vidual grains or lithic fragments and transported
to Rocknest from regional source areas.
Bulk Chemistry of the Rocknest
Sand Shadow
APXS provided an independent means of deter-
mining bulk chemistry of material in the Rock-
nest sand shadow. A measurement was made in
a wheel scuff named Portage, which was largely
devoid of surface crust (Fig. 1A). The chemical
composition (taking into account analytical un-
certainty) is within 2 SD of MER APXS analyses
of basaltic soils (Table 2). The APXS chem-
istry of basaltic soils analyzed by the MERs at
Gusev Crater and Meridiani Planum landing sites
(Table 2) are within 1 SD of each other except
for MgO and Na2O, which are the same within
2 SD (24–28). The MER compositional averages
exclude soils that contain a substantial local com-
ponent (high SO3 and high SiO2 for Gusev and
high Fe2O3 for Meridiani). The near identity of
compositions of the Rocknest, Gusev, and Merid-
ian basaltic soils implies either global-scale mix-
ing of basaltic material or similar regional-scale
basaltic source material or some combination
thereof.
Table 1. Mineralogy of Rocknest soil [CheMin x-ray diffraction (XRD)]
and normative mineralogies of basaltic materials from Gusev Crater
and of martian meteorites. (Rocknest data are amorphous-free values.)
Rocknest soil by CheMin (20), average of scoop 5, proportions of crystalline
phases normalized to 100%; values in italics uncertain. CIPW norms (weight) for
Gusev basaltic materials from MER APXS chemical analyses (26), ignoring S and
Cl; Fe3+
/Fetot for Backstay and Irvine taken as 0.17, the value for an Adirondack
basalt surface ground flat with the MER Rotary Abrasion Tool (RAT) (26). CIPW
norms (wt %) of martian meteorites from bulk compositions; Fe3+
/Fetot as
analyzed for Shergotty and Elephant Moraine (EETA) 79001A, estimated at
0.1 for Northwest Africa (NWA) 6234 and 0 for Queen Alexandra Range (QUE)
94201. K-spar is sanidine for the Rocknest soil, and normative orthoclase for
others. Low-Ca Pyx is pigeonite for the soil and normative hypersthene for
others. High-Ca Pyx is augite for the soil and normative diopside for others.
Fe-Cr oxide includes magnetite, hematite, and chromite. All phosphorus in
analyses are calculated as normative apatite. Mg no. is the % magnesium
substituting for iron in the olivine structure, An refers to the % Ca substituting
for Na in the plagioclase structure.
Location Gale Gusev Meteorites
Sample
Rocknest
sand shadow
Adirondack Backstay Irvine Shergotty
NWA
6234
EETA
79001A
QUE 94210
Quartz 1.4 0 0 0 0.2 0 0 3
Plagioclase 40.8 39 49 32 23 19 19 32
K-spar 1.3 1 6 6 1 0.5 0 0
Low-Ca Pyx 13.9 15 14 21 46 30 47 15
High-Ca Pyx 14.6 15 5 13 25 16 16 38
Olivine 22.4 20 15 16 0 27 13 0
Fe-Cr oxides 3.2 6 4 6 3 4 2 0
Ilmenite 0.9 1 2 2 2 2 1 4
Apatite – 1 3 2 2 2 1 6
Anhydrite 1.5
Mg no. 61 T 3 57 62 55 51 63 63 40
An 57 T 3 42 29 19 51 50 60 62
www.sciencemag.org SCIENCE VOL 341 27 SEPTEMBER 2013 1239505-3
RESEARCH ARTICLE

5. In contrast to the APXS measurement at
the Portage wheel scuff, both CheMin and SAM
measurements were carried out on the sieved,
<150-mm-size fraction of soil. To discriminate
potential differences between the fines deliv-
ered to CheMin and SAM and the bulk material
analyzed in the wheel scuff, APXS chemistry
was obtained from portions of sieved material
deposited on the observation tray. APXS spectra
from the bulk and sieved material are nearly iden-
tical, with the exception of a prominent Ti peak
and increased background from the observation
tray (reflecting Ti metal of the tray). Addition-
ally, Ca, Mn, and Fe signals in spectra from the
observation tray are lowered proportionally as
a function of their atomic number, which sug-
gests that a fraction of these grains is smaller
than the APXS sampling depth (29). Slightly ele-
vated S and Cl, with a S/Cl ratio similar to that
found in soils by MERs (30), suggest a potential
enrichment of these two elements in the <150-mm
fraction delivered to the observation tray.
To determine the amount and composition
of the amorphous component, mass balance cal-
culations were performed using the chemical
composition of the bulk sample, the chemical
compositions of the individual phases (e.g., pla-
gioclase, sanidine, and olivine) and the relative
proportions of those phases in the crystalline
component. The empirical formulas of the major
crystalline phases (Table 3) and their chemical
compositions (table S2) were calculated from
cell parameter data (20, 21) (table S1). The chem-
ical formulas and compositions of the minor
crystalline components were assigned by stoi-
chiometry (e.g., ilmenite as TiFeO3). The rela-
tive proportions of amorphous and crystalline
components and their respective bulk compo-
sitions are summarized in Table 4, with Rocknest
having ~45 weight percent (wt %) amorphous
and ~55 wt % crystalline components (31). The
chemical compositions and proportions of amor-
phous and crystalline components were calculated
on a light-element–free basis. The relative propor-
tion of the amorphous component will in reality
be greater than 45 wt % because the volatile in-
ventory is associated with that component (32).
Abundance estimates for the x-ray amorphous
component of a sample may vary considerably,
depending on the method used for their determi-
nation. Bish et al. (20), for example, used a full
pattern-fitting method together with known amor-
phous standard materials analyzed in the labo-
ratory to determine the amount of amorphous
or poorly crystalline material contained in the
CheMin x-ray diffraction pattern. Their reported
value of ~27 wt % T 50% (1 SD range of 13 to
40 wt %), as calculated from diffraction and
scattering data alone, is somewhat lower than
the ~45% calculated from mass balance consid-
erations, but both values are within the combined
analytical uncertainty of the two techniques.
The inferred chemical composition of the amor-
phous component (Table 4) contains ~23% FeO +
Fe2O3, suggesting that ferric nanophase oxide
[npOx (25, 26, 33)] is present in abundance.
Similarly, S (principally contained within the amor-
phous component) is closely associated with the
npOx in dunes at the MER sites (24, 27) as well.
Abundances of SO3 and Cl are correlated in soils
from Gusev and Meridiani, which implies that
both are associated with npOx in the amorphous
component because these elements are not asso-
ciated with Mg, Ca, or Fe in crystalline phases.
The elements Cr, Mn, and P were associated
with the amorphous component (Table 4), but
Table 2. Basaltic soil compositions from APXS analyses for Rocknest Portage, Gusev Crater,
and Meridiani Planum.
Rocknest Gusev Meridiani
Number 1* 48†
29†
SiO2 (wt %) 42.88 T 0.47 46.1 T 0.9 45.7 T 1.3
TiO2 1.19 T 0.03 0.88 T 0.19 1.03 T 0.12
Al2O3 9.43 T 0.14 10.19 T 0.69 9.25 T 0.50
Cr2O3 0.49 T 0.02 0.33 T 0.07 0.41 T 0.06
Fe2O3 + FeO 19.19 T 0.12 16.3 T 1.1 18.8 T 1.2
MnO 0.41 T 0.01 0.32 T 0.03 0.37 T 0.02
MgO 8.69 T 0.14 8.67 T 0.60 7.38 T 0.29
CaO 7.28 T 0.07 6.30 T 0.29 6.93 T 0.32
Na2O 2.72 T 0.10 3.01 T 0.30 2.21 T 0.18
K2O 0.49 T 0.01 0.44 T 0.07 0.48 T 0.05
P2O5 0.94 T 0.03 0.91 T 0.31 0.84 T 0.06
SO3 5.45 T 0.10 5.78 T 1.25 5.83 T 1.04
Cl 0.69 T 0.02 0.70 T 0.16 0.65 T 0.09
Br (mg/g) 26 T 6 53 T 46 100 T 111
Ni 446 T 29 476 T 142 457 T 97
Zn 337 T 17 270 T 90 309 T 87
Sum (wt %) 99.85 99.88 99.88
Cl/SO3 0.13 T 0.02 0.12 T 0.02 0.11 T 0.01
*Gellert et al., 2013 (35); analytical uncertainty. †T1SD of average.
Table 3. Empirical chemical formulas of the four
major phases identified in the Rocknest soil
estimated by crystal-chemical techniques.
Phase Formula
Olivine (Mg0.62(3)Fe0.38)2SiO4
Plagioclase (Ca0.57(13)Na0.43)(Al1.57Si2.43)O8
Augite (Ca0.75(4)Mg0.88(10)Fe0.37)Si2O6
Pigeonite (Mg1.13(9)Fe0.68(10)Ca0.19)Si2O6
Fig. 2. Pyroxene compositional quadrilateral, showing the chemical and thermal relations be-
tween the major igneous minerals in the Rocknest sand shadow. Compositions of augite, pigeonite,
and olivine in the Rocknest dune material, plotted on the pyroxene quadrilateral. En, enstatite, Mg2Si2O6;
Di, diopside, CaMgSi2O6; Hd, hedenbergite, CaFeSi2O6; and Fs, ferrosilite, Fe2Si2O6. Pyroxenes are plotted
within the quadrangle, based on CheMin XRD unit-cell parameters; olivine is plotted below the quad-
rilateral at the appropriate molar Mg/Fe ratio (20). Ellipses for each mineral approximate the uncer-
tainties in mineral compositions from their unit-cell parameters. Gray background lines represent the
surface of the pyroxene solvus, with temperatures in °C (40). Red lines are approximate equilibrium tie
lines from the augite centroid composition to compositions of olivine and pigeonite, based on similar
tie lines in an equilibrated anorthosite in lunar sample 62236 (41).
27 SEPTEMBER 2013 VOL 341 SCIENCE www.sciencemag.org1239505-4
Curiosity at Gale Crater

6. they could instead be present as crystalline phases
(e.g., Ca-phosphate and chromite) at abundances
below the CheMin detection limit and/or as sub-
stitutional impurities in the major crystalline phases
(e.g., Mn and Cr in pyroxene).
The SAM instrument analyzed Rocknest for
volatile species and organic molecules (32), and
it detected, in order of decreasing abundance,
H2O, SO2, CO2, and O2. The crystalline phases,
aside from a minor anhydrite component, do not
include these species as a part of their structure,
so they must either be present in the amorphous
component or be present in the crystalline com-
ponent at levels below the XRD detection limit,
or both.
ChemCam spot observations in the scoop
walls of Rocknest are characterized by the strong
emissions from elemental hydrogen, although
ChemCam is not sensitive to its bonding state (34).
Comparison of this result with those of CheMin
and SAM suggests that ChemCam detections
of hydrogen most likely correspond to the H2O
associated with the amorphous component de-
tected by CheMin.
Discussion
Global, Regional, and Local Sources
The crystalline phases in the Rocknest fines are
consistent with a basaltic source and fit well
within the measured qualitative mineralogy of
basaltic martian meteorites and the normative
mineralogy of Adirondack class olivine basalts
at Gusev Crater (25) (Table 1). If the Rocknest
assemblage of basaltic crystalline and amorphous
components is locally derived, it is distinct from
mafic float rocks analyzed to date by APXS and
ChemCam in Gale Crater (34, 35). This obser-
vation suggests that the similarity in the chem-
ical compositions of aeolian bedforms (basaltic
soil) at Gale, Gusev, and Meridiani (Table 2)
might result from global-scale aeolian mixing
of local-to-regional basaltic material that may
or may not have variable chemical composi-
tions. This process would require sufficiently
strong winds occurring with sufficient frequen-
cy over a long enough time to achieve global or
regional-scale transport of grains by saltation and
suspension.
An alternative explanation for the compara-
ble chemical compositions of aeolian bedforms
at Gale, Gusev, and Meridiani is that the chem-
ical compositions of martian basalts are similar
at regional scales everywhere on the planet. The
Rocknest sand shadow could reasonably have
locally sourced 1- to 2-mm particles, with finer-
grained regional basaltic material plus a contri-
bution from global dust. The similarity of soil
compositions (Table 2) suggests that the basaltic
fine-grained materials at Gusev, Meridiani, and
Gale Crater provide a reasonable approximation
to the bulk composition of the exposed martian
crust (36, 37).
It is tempting to suggest that the light-toned
martian dust is largely represented by the Rocknest
amorphous component. However, we have no
data to show that the <150-mm size fraction (clay
to fine-sand size fraction) of material analyzed
by CheMin has its finest material preferential-
ly enriched in amorphous material. The evi-
dence from MER for basaltic soils suggests that
the chemical composition of the fine-grained,
light-toned soil is approximately the same as the
coarser-grained, dark-toned soils [e.g., table 10
in (38)].
The central mound of Gale Crater (Mt. Sharp
or Aeolis Mons) exhibits reflectance spectra sug-
gesting the presence of crystalline hydrated sul-
fate minerals and phyllosilicates (39), but neither
was seen in Rocknest (above the 1 to 2% level).
The absence of material from Mt. Sharp could
arise from the wind pattern during formation
of the Rocknest sand shadow; it is oriented so
as to imply sediment transport from the north,
and Mt. Sharp is east and southeast of Rocknest.
Materials and Methods
Calculation of Wind Speeds Required
to Form the Rocknest Sand Shadow
The wind velocity required to move the coarse
grains of the sand shadow by creep can be cal-
culated. The critical shear velocity (u*c) of the
wind needed to transport 1-mm-diameter (d) grains
is given by (42) as
u*c ¼
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
0:0123 sgd þ
0:0003 kg=s2
˜nf d
!v
u
u
t
where s ¼ ˜ns=˜nf , ˜ns is the density of the grains
using basalt (3000 kg/m3
), ˜nf is the density of
Table 4. Chemical composition and proportion of XRD amorphous component in Rocknest Portage from APXS and CheMin data.
Origin Remove XRD crystalline component* Composition
APXS† APXS+
CheMin
Plagio-
clase
San-
idine
Olivine Augite
Pigeon-
ite
Ilmen-
ite
Hema-
tite
Mag-
netite
Anhy-
drite
Quartz
Amor-
phous‡
Crystal-
line
SiO2, wt % 42.88 42.88 30.88 30.42 25.95 21.63 17.51 17.51 17.51 17.51 17.51 16.76 37.20 47.59
TiO2 1.19 1.19 1.19 1.19 1.19 1.19 1.19 0.93 0.93 0.93 0.93 0.93 2.06 0.47
Al2O3 9.43 9.43 2.85 2.72 2.72 2.72 2.72 2.72 2.72 2.72 2.72 2.72 6.04 12.24
Cr2O3 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 1.09 0.00
FeO+Fe2O3
§
19.19 10.43 10.43 10.43 10.43 10.43 10.43 10.43 10.43 10.43 10.43 10.43 23.14 -0.10
FeO-Cryst||
— 7.37 7.37 7.37 3.31 2.29 0.59 0.35 0.35 0.00 0.00 0.00 -0.01 13.48
Fe2O3-Cryst¶
— 1.39 1.39 1.39 1.39 1.39 1.39 1.39 0.79 0.00 0.00 0.00 -0.01 2.55
MnO 0.41 0.41 0.41 0.41 0.41 0.41 0.41 0.41 0.41 0.41 0.41 0.41 0.91 0.00
MgO 8.69 8.69 8.69 8.69 4.97 3.72 2.19 2.19 2.19 2.19 2.19 2.19 4.86 11.86
CaO 7.28 7.28 4.65 4.65 4.65 3.19 2.87 2.87 2.87 2.87 2.53 2.53 5.61 8.67
Na2O 2.72 2.72 1.62 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 1.60 3.56 2.03
K2O 0.49 0.49 0.49 0.40 0.40 0.40 0.40 0.40 0.40 0.40 0.40 0.40 0.89 0.16
P2O5 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 0.94 2.09 -0.01
SO3 5.45 4.96 4.96 4.96 4.96 4.96 4.96 4.96 4.96 4.96 4.96 4.96 11.01 -0.05
SO3-Cryst#
— 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.49 0.00 0.00 -0.01 0.90
Cl 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.61 0.61 1.35 -0.01
Sum 99.77 99.77 77.47 76.77 64.52 56.47 48.80 48.30 47.70 46.55 45.71 44.96 99.77 99.77
∑(FeO+Fe2O3) 19.19 19.19 — — — — — — — — — — 23.14 16.03
∑(SO3) 5.54 5.54 — — — — — — — — — — 11.01 0.90
Relative to whole sample 22.3 0.7 12.3 8.0 7.6 0.5 0.6 1.2 0.8 0.8 45.3 54.7
Relative to XRD crystalline 40.8 1.3 22.4 14.6 13.9 0.9 1.1 2.1 1.5 1.4 — 100.0
*Plagioclase, An57; Olivine, Fo62; Augite, En44Fs20Wo36 (Mg/Fe, 2.2 atomic); Pigeonite, En56Fs35Wo8 (Fe/Mg, 1.6 atomic). †APXS chemistry from Gellert et al. (35). ‡Cr2O3 and
MnO calculated with the amorphous component. §Total Fe as FeO+Fe2O3 because APXS does not distinguish oxidation states. ||FeO required for Fe2+
crystalline phases (olivine,
augite, pigeonite, ilmenite, and magnetite). ¶Fe2O3 required for Fe3+
crystalline phases (hematite and magnetite). #SO3 required for crystalline SO3 crystalline phase (anhydrite).
www.sciencemag.org SCIENCE VOL 341 27 SEPTEMBER 2013 1239505-5
RESEARCH ARTICLE

7. martian air (0.02 kg/m3
), and g is the acceleration
due to gravity (3.71 m/s2
). The calculated u*c is
2.6 m/s, which represents the fluid shear veloc-
ity to initiate motion. Because grains in creep
derive a portion of their momentum from colli-
sions by saltating grains, on Earth once saltation
begins, creep can occur down to 0.7 u*c (1.8 m/s
as applied to the Rocknest grains), which repre-
sents the impact threshold for motion. Given a
boundary layer created by winds blowing over
the surface, shear velocities can then be related
to the wind speeds above the surface by the law
of the wall
uz ¼
u*
k
ln
z
z0
where uz is the wind speed at height z above the
surface (taken here as 1 m), k is a constant of
0.407, and z0 is the roughness height where the
idealized logarithmic wind profile is predicted to
be zero. Roughness height varies by grain size
and the height of surface features, such as wind
ripples (7), and also by the height and intensity
of the saltation cloud (43). Rocknest conditions
are unknown, but z0 is taken as 0.3 mm, which
would be the roughness height with wind rip-
ples 10 mm in height. Estimated wind speeds
at 1 m above the surface are ~52 m/s (116 mph)
and 36 m/s (80 mph), without and with saltation,
respectively. As a result of the lower gravity and
reduced atmospheric density on Mars, a greater
hysteresis exists than on Earth between the fluid
and impact thresholds, and saltation impacts upon
grains are more energetic (18, 44, 45). The com-
bined effects suggest that initial transport of the
coarse surface grains probably occurred at lower
wind speeds than those calculated. Conversely,
reactivation of the sand shadow would require
considerably higher wind speeds because of in-
duration of the surface.
Although observations from the Viking Lander
1 suggest that wind speeds of 30 m/s at a height
of 1.6 m occurred during its 2-year lifetime (46),
we do not known how often Mars winds can be
capable of transporting 1- to 2-mm grains. The
wind estimates above suggest that formation
of the Rocknest sand shadow has involved rare
strong winds and that reactivation of the sand
shadow from its currently indurated state would
require even stronger and rarer winds.
Given the possibility of considerable antiquity
of the Rocknest sand shadow and similar coarse-
grained bedforms on Mars, could their activa-
tion correspond to the martian obliquity cycle?
At low obliquities, the atmosphere collapses onto
the polar caps, but at high obliquity, CO2 is re-
leased to the atmosphere (47, 48). Taken as an
end member, atmospheric density may double at
high obliquity and thereby enhance aeolian ac-
tivity (48). As a comparison with the above val-
ues calculated for the present martian atmosphere,
using 0.04 kg/m3
for atmospheric density, the
calculated fluid u*c is 1.9 m/s and the impact u*c
is 1.3 m/s, which correspond to wind speeds at
the 1-m height of ~37 m/s (83 mph) and 26 m/s
(58 mph), respectively. Although considerably
lower than values calculated for present condi-
tions, rare strong wind events are still implied.
References and Notes
1. A Mars solar day has a mean period of 24 hours, 39 min,
35 s and is customarily referred to as a “sol” to
distinguish it from the roughly 3% shorter day on Earth.
2. A sand shadow is an accumulation of wind-blown sediment
deposited in the lower-velocity lee of an obstacle in
the path of the wind.
3. R. C. Anderson et al., Collecting samples in Gale Crater,
Mars; An overview of the Mars Science Laboratory Sample
Acquisition, Sample Processing and Handling System.
Space Sci. Rev. 170, 57–75 (2012). doi: 10.1007/
s11214-012-9898-9
4. M. S. Anderson et al., In situ cleaning of instruments
for the sensitive detection of organics on Mars. Rev.
Sci. Instrum. 83, 105109 (2012). doi: 10.1063/1.4757861;
pmid: 23126806
5. D. F. Blake et al., Characterization and calibration of the
CheMin mineralogical instrument on Mars Science
Laboratory. Space Sci. Rev. 170, 341–399 (2012).
doi: 10.1007/s11214-012-9905-1
6. P. R. Mahaffy et al., The sample analysis at Mars
investigation and instrument suite. Space Sci. Rev. 170,
401–478 (2012). doi: 10.1007/s11214-012-9879-z
7. R. A. Bagnold, The Physics of Blown Sand and Desert
Dunes (Chapman and Hall, London, 1941).
8. K. S. Edgett et al., Curiosity’s Mars Hand Lens Imager
(MAHLI) Investigation. Space Sci. Rev. 170, 259–317
(2012). doi: 10.1007/s11214-012-9910-4
9. W. Goetz et al., Microscopic analysis of soils at the
Phoenix landing site, Mars: Classification of soil
particles and description of their optical and magnetic
properties. J. Geophys. Res. 115, E00E22 (2010).
doi: 10.1029/2009JE003437
10. K. E. Herkenhoff et al., In situ observations of the
physical properties of the martian surface, in The Martian
Surface: Composition, Mineralogy, and Physical
Properties, J. F. Bell III, Ed. (Cambridge Univ. Press,
Cambridge, 2008), pp. 451–467.
11. R. Sullivan et al., Wind-driven particle mobility on Mars:
Insights from Mars Exploration Rover observations at
“El Dorado” and surroundings at Gusev Crater. J. Geophys.
Res. 113, E06S07 (2008). doi: 10.1029/2008JE003101
12. L. A. Soderblom et al., Soils of Eagle Crater and Meridiani
Planum at the Opportunity rover landing site. Science
306, 1723–1726 (2004). doi: 10.1126/science.1105127;
pmid: 15576606
13. R. Sullivan et al., Aeolian processes at the Mars
exploration rover Meridiani Planum landing site. Nature
436, 58–61 (2005). doi: 10.1038/nature03641;
pmid: 16001061
14. S. G. Fryberger, P. Hesp, K. Hastings, Aeolian granule
ripple deposits, Namibia. Sedimentology 39, 319–331
(1992). doi: 10.1111/j.1365-3091.1992.tb01041.x
15. D. J. Jerolmack, D. Mohrig, J. P. Grotzinger, D. A. Fike,
W. A. Watters, Spatial grain size sorting in eolian ripples
and estimation of wind conditions on planetary surfaces:
Application to Meridiani Planum, Mars. J. Geophys. Res.
111, E12S02 (2006). doi: 10.1029/2005JE002544
16. J. M. Ellwood, P. D. Evans, I. G. Wilson, Small scale
aeolian bedforms. J. Sed. Petrol. 45, 554–561 (1975).
17. P. A. Hesp, The formation of shadow dunes. J. Sed. Petrol
51, 101–112 (1981).
18. M. P. Almeida, E. J. R. Parteli, J. S. Andrade Jr.,
H. J. Herrmann, Giant saltation on Mars. Proc. Natl.
Acad. Sci. U.S.A. 105, 6222–6226 (2008). doi: 10.1073/
pnas.0800202105; pmid: 18443302
19. M. P. Golombek et al., Constraints on ripple migration at
Meridiani Planum from Opportunity and HiRISE
observations of fresh craters. J. Geophys. Res. 115,
E00F08 (2010). doi: 10.1029/2010JE003628
20. D. L. Bish et al., X-Ray diffraction results from Mars
Science Laboratory: Mineralogy of Rocknest at Gale
Crater. Science 341, 1238932 (2013); doi: 10.1126/
science.1238932
21. Supplementary materials are available on Science
Online.
22. Unit cell parameters obtained from the RRUFF Project
database, http://rruff.info/ima.
23. S. R. Taylor, S. M. McLennan, Planetary Crusts:
Their Composition, Origin and Evolution (Cambridge
Univ. Press, Cambridge, (2009).
24. R. V. Morris et al., Iron mineralogy and aqueous
alteration from Husband Hill through Home Plate at
Gusev Crater, Mars: Results from the Mössbauer
instrument on the Spirit Mars Exploration Rover.
J. Geophys. Res. 113, E12S42 (2008).
doi: 10.1029/2008JE003201
25. D. W. Ming et al., Geochemical properties of rocks and
soils in Gusev Crater, Mars: Results of the Alpha Particle
X-ray Spectrometer from Cumberland Ridge to Home
Plate. J. Geophys. Res. 113, E12S39 (2008).
doi: 10.1029/2008JE003195
26. R. V. Morris et al., Mössbauer mineralogy of rock,
soil, and dust at Gusev Crater, Mars: Spirit’s journey
through weakly altered olivine basalt on the Plains
and pervasively altered basalt in the Columbia Hills.
J. Geophys. Res. 111, E02S13 (2006).
doi: 10.1029/2005JE002584
27. A. S. Yen et al., An integrated view of the chemistry
and mineralogy of martian soils. Nature 436, 49–54
(2005). doi: 10.1038/nature03637; pmid: 16001059
28. A. S. Yen et al., Evidence for a global martian soil
composition extends to Gale Crater. 45th Lunar and
Planetary Science Conference, March 2013, Published
on CD by the Lunar and Planetary Institute, Houston,
Texas, Abstract 2495 (2013).
29. J. A. Berger et al., MSL Titanium Observation Tray
Measurements with APXS. 45th Lunar and Planetary
Science Conference, March 2013, Published on CD
by the Lunar and Planetary Institute, Houston, Texas,
Abstract 1321 (2013).
30. R. Gellert et al., Alpha Particle X-ray Spectrometer
(APXS): Results from Gusev Crater and calibration
report. J. Geophys. Res. 111, E02S05 (2006).
doi: 10.1029/2005JE002555
31. Because APXS does not discriminate among iron oxidation
states, the total Fe concentration was proportioned in
accordance with the oxidation state information carried by
the crystalline phases (Table 3, column 3). FeO-Cryst and
Fe2O3-Cryst are the concentrations of FeO and Fe2O3
required to accommodate olivine, augite, pigeonite,
ilmenite, and magnetite and hematite, in accordance with
their valence states. The remaining iron (FeO + Fe2O3) is
then associated with the amorphous component without
implications for oxidation state. Similarly, some SO3 is
reported as SO3-Cryst to accommodate anhydrite as a
crystalline component.
32. L. A. Leshin et al., Volatile, isotope, and organic
analysis of martian fines with the Mars Curiosity Rover.
Science 341, 1238937 (2013); doi: 10.1126/
science.1238937
33. Nanophase ferric oxide (npOx) is a generic name for
amorphous, poorly crystalline, or short-range ordered
products of oxidative alteration/weathering that have
octahedrally coordinated Fe3+
(Mössbauer doublet)
and are predominantly oxide/oxyhydroxide/hydrous in
nature. Depending on local conditions, npOx (as
encountered on Earth) can be any combination of
superparamagnetic hematite and goethite, lepidocrocite,
ferrihydrite, schwertmannite, akaganeite, hisingerite,
and the octahedral Fe3+
-rich particles that pigment
iddingsite and palagonite. npOx can also incorporate
anions like (SO4)2–
, Cl–
, and (PO4)3–
through specific
chemical adsorption. Because of different local conditions
on Mars, one or more forms of npOx on the planet
may be uncommon or not present on Earth.
34. P.-Y. Meslin et al., Soil diversity and hydration as observed
by ChemCam at Gale Crater, Mars. Science 341,
1238670 (2013); doi: 10.1126/science.1238670
35. R. Gellert et al., Initial MSL APXS activities and
observations at Gale Crater, Mars, 45th Lunar and
Planetary Science Conference, March 2013, Published
on CD by the Lunar and Planetary Institute, Houston,
Texas, Abstract 1432 (2013).
27 SEPTEMBER 2013 VOL 341 SCIENCE www.sciencemag.org1239505-6
Curiosity at Gale Crater

8. 36. H. Y. McSween Jr., G. J. Taylor, M. B. Wyatt, Elemental
composition of the martian crust. Science 324, 736–739
(2009). doi: 10.1126/science.1165871; pmid: 19423810
37. S. R. Taylor, S. M. McLennan, Planetary Crusts:
Their Composition, Origin and Evolution (Cambridge
Univ. Press, Cambridge, 2009).
38. R. V. Morris et al., Mössbauer mineralogy of rock, soil,
and dust at Meridiani Planum, Mars: Opportunity’s
journey across sulgate-rich outcrop, basaltic sand and
dust, and hematite lag deposits. J. Geophys. Res. 111,
E12S15 (2006). doi: 10.1029/2005JE002584
39. R. E. Milliken, J. P. Grotzinger, B. J. Thomson,
Paleoclimate of Mars as captured by the stratigraphic
record in Gale Crater. GRL 37, L04201 (2010).
doi: 10.1029/2009GL041870
40. D. H. Lindsley, Pyroxene thermometry. Am. Mineral. 68,
477–493 (1983).
41. P. H. Warren, J. T. Wasson, The compositional-petrographic
search for pristine nonmare rocks: Third foray. Proc. Lunar
Planet. Sci. Conf. 10th (1979), 583–610.
42. Y. Shao, H. Lu, A simple expression for wind erosion
threshold friction velocity. J. Geophys. Res. 105, (D17),
22,437–22,443 (2000). doi: 10.1029/2000JD900304
43. P. R. Owen, Saltation of uniform grains in air. J. Fluid Mech.
20, 225–242 (1964). doi: 10.1017/S0022112064001173
44. P. Claudin, B. Andreotti, A scaling law for aeolian
dunes on Mars, Venus, Earth, and for subaqueous
ripples. Earth Planet. Sci. Lett. 252, 30–44 (2006).
doi: 10.1016/j.epsl.2006.09.004
45. J. F. Kok, Difference in the wind speeds required for
initiation versus continuation of sand transport on Mars:
Implications for dunes and dust storms. Phys. Rev. Lett.
104, 074502 (2010). doi: 10.1103/PhysRevLett.104.074502;
pmid: 20366891
46. R. E. Arvidson, E. A. Guinness, H. J. Moore, J. Tillman,
S. D. Wall, Three Mars years: Viking Lander 1 imaging
observations. Science 222, 463–468 (1983). doi: 10.1126/
science.222.4623.463; pmid: 17746178
47. C. E. Newman, S. R. Lewis, P. L. Read, The atmospheric
circulation and dust activity in different orbital epochs on
Mars. Icarus 174, 135–160 (2005). doi: 10.1016/
j.icarus.2004.10.023
48. R. J. Phillips et al., Massive CO2 ice deposits sequestered
in the south polar layered deposits of Mars. Science 332,
838–841 (2011). doi: 10.1126/science.1203091;
pmid: 21512003
Acknowledgments: Support from the NASA Mars Science
Laboratory Mission is gratefully acknowledged. The chemical
and mineralogical data presented here are derived from the
archived data sets in the NASA Planetary Data System (PDS)
http://pds-geosciences.wustl.edu/missions/msl, specifically
MSL-M-CHEMIN-2-EDR-V1.0 and MSL-M-APXS-2-EDR-V1.0.
M.B.M. was funded by the Danish Council for Independent
Research/Natural Sciences (Det Frie Forskningsråd Natur og
Univers FNU grants 12-127126 and 11-107019).
W.G. acknowledges partial funding by the Deutsche
Forschungsgemeinschaft (DFG grant GO 2288/1-1).
Some of this research was carried out at the Jet Propulsion
Laboratory, California Institute of Technology, under a
contract with NASA.
Supplementary Materials
www.sciencemag.org/content/341/6153/1239505/suppl/DC1
Supplementary Text
Figs. S1 to S4
Tables S1 and S2
References
23 April 2013; accepted 31 July 2013
10.1126/science.1239505
www.sciencemag.org SCIENCE VOL 341 27 SEPTEMBER 2013 1239505-7
RESEARCH ARTICLE